Western Michigan University

ScholarWorks at WMU

Honors Theses Lee Honors College

4-16-2024

Design of Telemetry and Remote-Control System for Electric

Propulsion Testing
Logan Alvesteffer
Western Michigan University

Follow this and additional works at: [Link]

Part of the Aerospace Engineering Commons

Recommended Citation
Alvesteffer, Logan, "Design of Telemetry and Remote-Control System for Electric Propulsion Testing"
(2024). Honors Theses. 3784.
[Link]

This Honors Thesis-Open Access is brought to you for

free and open access by the Lee Honors College at
ScholarWorks at WMU. It has been accepted for inclusion
in Honors Theses by an authorized administrator of
ScholarWorks at WMU. For more information, please
contact wmu-scholarworks@[Link].
 Design of Telemetry and Remote-Control System for Electric Propulsion Testing

Logan Alvesteffer
Douglas Adams
Adam Tuckey

Western Michigan University

Kalamazoo, MI 49008

Dr. Kristina Lemmer and Dr. Bade Shrestha

April 19, 2024

Disclaimer
This project report was written by students at Western Michigan University to fulfill an engineering
curriculum requirement. Western Michigan University makes no representation that the material
contained in this report is error-free or complete in all respects. Persons or organizations who
choose to use this material do so at their own risk.
 Abstract
Ground testing of electric propulsion devices utilize high-frequency and low-frequency
telemetry systems to provide critical operational information about the device’s performance.
These values include thruster operating parameters like voltage, current, and gas flow rate, while
simultaneously reporting vacuum chamber pressure and temperature. This work encompasses the
design, construction, and validation of a modular telemetry and remote-control system for the
testing of electric propulsion devices in Western Michigan University’s Aerospace Laboratory for
Plasma Experiments. The integrated telemetry system facilitates simultaneous control and data
collection across laboratory subsystems, such as power supplies and thruster-specific
measurements. These systems are integrated through modular breakout boxes to allow for agnostic
thruster operation and selection within the Cold Plasma Experimental Research Station. Validation
of these designs utilized, or will utilize, the monitoring of low power WHTms-44 Hall effect
thruster and high-current H6 hollow cathode experiments to confirm agnostic capabilities.

1
 Table of Contents
1. Motivation & Background 7
1.1 Discussion of Electric Propulsion 7
1.2 Importance of Electric Propulsion: Psyche Mission 10
1.3 Design Objectives 11
1.4 Agnostic Capabilities of CEReS 11
2. System Design 13
2.1 Telemetry Break-out-Box 14
2.1.1 CAD and Simulations 14
2.1.2 Final Design 17
2.2 Discharge Filter and Power Control Break-out-Box 19
2.2.1 CAD and Simulations 19
2.2.2 Final Design 21
2.3 Current & Voltage Sensors 22
2.3.1 Telemetry Current Sensor Trade Study 22
2.3.2 Telemetry Voltage Divider Design 27
2.4 Overall Design and Layout 31
3. Remote Control and Monitoring 32
3.1 Serial Connections 32
3.1.1 ALICAT Mass Flow Controller 32
3.1.2 Lambda GEN Power Supplies 33
3.2 Analog Connections 33
3.2.1 Remote Monitoring 34
[Link] Power supplies 34
[Link] Gate Valves 35
3.2.2 Analog Remote Control 36
[Link] Circuit design 37
[Link] Simulations 40
[Link] Bill of Materials 41
[Link] Additional Considerations 41
3.3 LabVIEW Virtual Instrument 42
3.3.1 Front Panel Layout 42
3.3.2 Block Diagram Structure 45
4. Sensor Benchmarking 46
4.1 Simulated Load Tests 46
4.2 H6 Hollow Cathode Tests 48
5. Societal Impacts 50
6. Conclusions 51
7. Future Recommendations 51
Acknowledgements 52
References 53
Appendix A: Decision Matrix 54
Appendix B: Digital Resolution MATLAB Code 55
Appendix C: SolidWorks Simulations 56
Appendix D: IEPC Abstract 57
Appendix E: ABET Questionnaire 59

2
 Table of Figures
Figure 1. Ion Thruster Diagram (Goebel and Katz) 8
Figure 2. Hall-Effect Thruster Diagram (Goebel and Katz) 8
Figure 3. Hollow Cathode Diagram (Goebel and Katz) 9
Figure 4. Psyche Mission 10
Figure 5. ALPE 11
Figure 6. Front & Back CEReS Power Rack Views 13
Figure 7. Telemetry BoB CAD Model 14
Figure 8. Telemetry BoB Power Ratings 16
Figure 9. Circuitry Layout for Telemetry BoB Trays 17
Figure 10. Telemetry BoB Tray 17
Figure 11. Telemetry BoB Prototype 18
Figure 12. Prior Discharge System Arrangement 19
Figure 13. CAD Model of Discharge Filter & Power Control BoB 20
Figure 14. Discharge Filter Banana Connections 21
Figure 15. Discharge Filter Prototype 22
Figure 16. Custom CQ330 Hall Sensor Circuit Schematic 23
Figure 17. Custom CQ330 Hall Sensor PCB 23
Figure 18. Technical Drawing of the Verivolt IsoBlock Sensor 24
Figure 19. FluxGate Methodology 24
Figure 20. Shunt-Based Methodology 25
Figure 21. Technical Drawing of the VEXTREX VCS40 26
Figure 22. Hall Effect Methodology 26
Figure 23. Average Time Current Curves of 216 Series (left) & 0ADBP & 0ADAP Series (right) Fuses 28
Figure 24. Voltage Divider Schematic 30
Figure 25. Voltage Divider PCB Layout 30
Figure 26. Rack Arrangement 31
Figure 27. ALICAT Mass Flow Controller 32

3
Figure 28. Lambda GEN-model Power Supply Unit 33
Figure 29. Analog Communication Flow Chart 33
Figure 30. Power Supply Measurement Flow Chart 34
Figure 31. Gate Valve Status Measurement Circuitry 35
Figure 32. Lambda EMS-model Power Supply Units 36
Figure 33. Analog Control System Design Flow Chart 37
Figure 34. Pulse-width Modulated Digital Signal Duty Cycle 38
Figure 35. Pulse-with Modulated (left) to Analog Signal (right) Conversion 38
Figure 36. First Order Lowpass Filter 39
Figure 37. Simulated LTSpice Circuits 40
Figure 38. Simulated LTSpice Results of Voltage Supply (left) and Current Supply (right) 40
Figure 39. LabVIEW Telemetry Panel 42
Figure 40. LabVIEW Control Panel 43
Figure 41. LabVIEW Configure Panel 44
Figure 42. LabVIEW Instructions Panel 45
Figure 43. Experimental Load Setup 46
Figure 44. Sensor Benchmarking: Current Sensor Load Tests 47
Figure 45. Sensor Benchmarking: Voltage Divider Load Tests 47
Figure 46. H6 Testing 48
Figure 47. H6 Testing: VCS40 vs Pearson Coil, Non-Normalized, Magnetized 49
Figure 48. H6 Testing: VCS40 vs Pearson Coil, Normalized, Magnetized 49
Figure 49. H6 Testing: VCS40 vs Pearson Coil, Normalized, Nonmagnetized 49

4
 Table of Tables
Table 1. Variable Thruster Data 7
Table 2. Power Supply Information 15
Table 3. Trade Study of Current Sensor Methodology 27
Table 4. Voltage Divider Design 1 28
Table 5. Voltage Divider Design 2 29
Table 6. Voltage Divider Fuse Values 29
Table 7. Circuit Component Output Full Scale Range and Digital Resolution 35
Table 8. Analog Power Supply Control System Component Costs 41

5
 Nomenclature
ALPE –Aerospace Laboratory for Plasma Experiments
BoB – Breakout Box
CEReS – Cold plasma Experimental Research Station
DAQ – Data Aquisition Unit
ECR – Electron Cyclotron Resonance
EP – Electric Propulsion
FS – Full Scale
FSR – Full Scale Range
HET – Hall Effect Thruster
PWM – Pulse-width Modulation/Modulated
RF – Radio Frequency
WMU – Western Michigan University

6
1. Motivation and Background

1.1 Discussion of Electric Propulsion

EP refers to the use of electrical energy to propel spacecraft. Unlike traditional chemical
propulsion systems that rely on the combustion of chemical propellants, electric propulsion
systems use electric power to accelerate charged particles, ions and/or electrons, to generate thrust.
Electric propulsion systems typically use electrostatic or electromagnetic effects to ionize and
accelerate propellant to high velocities, which are expelled at high speeds to produce thrust. Noble
gases such as xenon, argon, and krypton are typically used as propellants. Furthermore, for
instance, electrostatic thrusters, commonly known as ion thrusters, are comprised of an ionization
cathode, a series of charged grids to accelerate the ions, and a neutralizer cathode to emit electrons,
neutralizing the plasma plume. Electrostatic thrusters are different from electromagnetic thrusters
in that they exclusively use an electric field, instead of an electric and magnetic field to produce
thrust. The table below shows performance figures for various propulsion systems with flight
heritage, and the following figures depict examples of both electrostatic and electromagnetic EP
devices [1-6].

Table 1. Variable Thruster Data

Specific Impulse Input Power Efficiency
Thruster Propellant
(s) (kW) Range (%)
Cold Gas 50-75 - - Various
Chemical
150-225 - - ,
(Monopropellant)
Chemical
300-450 - - Various
(Bipropellant)
Resistojet 300 0.5-1 65-90
Arcjet 500-600 0.9-2.2 25-45
Ion Thruster 2500-3600 0.4-4.3 40-80 Noble Gas
Hall Thruster 1500-2000 1.5-4.5 35-60 Noble Gas

7
 Figure 1. Ion Thruster Diagram (Goebel and Katz)

Figure 2. Hall-Effect Thruster Diagram (Goebel and Katz)

8
 Several common types of electrostatic EP devices include HETs, pictured in Figure 3, ECR,
and RF thrusters. These different EP systems have various advantages over one another. Most types
of EP thrusters have much higher specific impulses than chemical propulsion systems, which prove
them useful for orbital station-keeping maneuvers, interplanetary missions, and deep space
missions. One of the most prevalent types of electrostatic propulsion is the HET, which uses the
Hall-effect to both ionize and expel plasma from the thruster at high speeds. Electrons, emitted by
a cathode, travel along electric and magnetic field lines towards the channel's interior.
These electrons collide with propellant atoms, creating ionization events. The ionized gas is then
accelerated away from the anode and expelled from the discharge channel at great speed, creating
thrust. Downstream of the thruster, the ionized gas mixes with electrons not drawn into the channel,
creating the quasi-neutral plasma plume [1-6[.
Another common EP device is the hollow cathode, which is used to supply electrons for
propellant ionization and beam neutralization. Hollow cathode operation is driven by thermionic
electron emission, which is the discharge of electrons from heating the insert material. This type
of cathode mainly emits electrons and is commonly used in ion thrusters and HETs. Conventional
hollow cathodes use an impregnated insert that releases electrons when heated by an external
resistive heater. Noble gases interact with the free electrons and ionize, creating plasma; this
plasma is extracted using an electric field and gas pressure. Hollow cathodes are heated internally
after the ionization reaches a critical threshold; the external heater is deactivated during operation
and the cathode insert is then heated to emission through plasma bombardment. This heating
occurs either through orifice, ion, or electron heating techniques. A diagram of a common hollow
cathode is shown below in figure 4, where the insert and orifice plate are shown [1-6].

Figure 3. Hollow Cathode Diagram (Goebel and Katz)

9
1.2 Importance of Electric Propulsion: Psyche Mission
Spacecraft EP systems are becoming more prevalent year after year due to their reliability,
cost of operation, propellant mass savings, and overall efficiency when compared to traditional
chemical combustion. One such example can be seen with the recent launch of NASA’s Psyche
Mission, where a set of four onboard HETs at 5kW and ~0.06 lbs of thrust each is propelling this
spacecraft to the asteroid belt. This journey is to a unique metal-rich asteroid orbiting the Sun
between Mars and Jupiter, which is named Psyche, after the Greek goddess of the soul. This
asteroid is of interest to NASA due to its possibly exposed nickel-iron core of an early planet,
hence the name. This mission may offer a unique window into the history of celestial collisions
and accretions which created terrestrial planets, such as Earth and its theorized metallic core. The
onboard HET system will make efficient and precise maneuvers over 3 years to reach Psyche and
conduct scientific operations. The inclusion of HETs allows the Psyche spacecraft to carry more
research and scientific instruments and equipment than a chemically propelled satellite, while
maintaining both launch and operation cost savings [7].

Figure 4. Psyche Mission

10
1.3 Design Objectives
The purpose of this design project was to design a telemetry system for ALPE’s CEReS
vacuum chamber to be utilized during EP experiments. This required the upgrading of prior NASA
Glenn LabVIEW programs and HET discharge filter configurations to be propulsion system
agnostic. Design and construction of diagnostic, protective, and remote-control hardware provides
active monitoring of the power supplies and related CEReS subsystems. CEReS hosts a variety of
EP experiments which required agnostic design considerations dependent on power supply, DAQ,
and remote-control parameters. The overall design requires modular and cost-effective hardware
and software systems that can be integrated and operated based on varying equipment and EP
device parameters. Referencing Appendix A, the decision-matrix highlights the individual
decisions necessary during the conceptual and preliminary telemetry and remote-control design.
These include Break-out Box positioning, Break-out Box design, current sensor methodology,
circuit design, and LabVIEW integration [5-6].

1.4 Agnostic Capabilities of CEReS

Figure 5. ALPE

11
 WMU’s ALPE, pictured above in Figure 5, investigates plasma experiments and diagnostic
devices for space propulsion, enhanced combustion, and environmental testing. CEReS, the main
vacuum chamber in ALPE, is outfitted with a CTI-250F and CTI-10 cryogenic pump for a
combined pumping capacity of 5,200 L/s-air. CEReS facilitates the testing of various EP devices,
including a high-power HET, a low-power hollow cathode, and a low-current heaterless hollow
cathode. These devices have diverse voltage and current requirements, which require necessary
protective circuitry for their operation. For example, the HET utilizes an integrated discharge filter
due to the oscillatory nature of HETs, and the heaterless hollow cathode requires ballast resistors
and protective diodes.
The current telemetry system of CEReS is only configured for HET testing. Therefore, it
is extremely inconvenient for CEReS to host ground tests of hollow cathodes and other EP devices
with different requirements. These tests are performed, but the process of telemetry equipment
interchange is tedious and time consuming. Additionally, the telemetry system of CEReS does not
have instrument remote control or direct-record monitoring capability. Thus, researchers must
control and monitor all hardware manually, reducing the accuracy of measurements. To increase
the capabilities of ALPE and allow for quick-turnaround testing of a variety of EP devices with
minimal effort, an agnostic telemetry and remote-control system was designed, capable of variable
and selectable circuitry, allowing for seamless functionality while accounting for the variations
between thruster devices.

12
2. System Design

The CEReS test station is equipped with seven power supplies that vary in power output.
The telemetry system is designed to monitor these outputs specific to each power supply using
integrated circuitry. Voltage dividers and FluxGate current sensors provide time-averaged
measurements, while pearson coils and Vektrex current sensors and attachable high-voltage probes
provide time-resolved measurements. With the exception of the high-voltage probes, the diagnostic
devices are housed inside plexiglass BoBs and mounted into the power rack behind the power
supplies. Measurements taken by the telemetry system are collected using National Instruments’
USB-6229 DAQ system and recorded/analyzed using a LabVIEW software suite.
Figure 6 shows how the modular BoBs are mounted onto the rack. BoB components are
categorized by their function; current/voltage measurements are handled separately from power
control, discharge protection, and pressure measurement. The power control system routes the
power input and grounding for the power supplies and additional hardware. The protection and
ignition system implements a discharge filter for HETs and blocking and feedback diodes for
heaterless hollow cathode ignition. The voltage/current measurement and monitoring system house
time-averaged and time-resolved diagnostic hardware. The power control system allows for
computer control of power supply output configurations. The pressure monitoring system
incorporates measurements compiled by a B-RAX pressure monitor into the LabVIEW suite.
These systems are arranged in a cascading configuration, which will allow for the connections to
be easily repositioned depending on EP device selection. Thermal modeling was performed, and a
ducting system is integrated to ensure thermal protection of components. The individual BoBs link
directly to the DAQ installed in the rack, where the LabVIEW program captures the time-averaged
measurements. The LabVIEW program also initiates data collection from external oscilloscopes
to capture time-resolved measurements. This LabVIEW program also facilitates remote operation
of power supplies, vacuum chamber gate valves, and mass flow controllers.

Figure 6. Front & Back CEReS Power Rack Views

13
2.1 Telemetry Break-out-Box
2.1.1 CAD & Simulations
The telemetry-gathering section of the BoB is designed to provide two measurements for each
connected power supply, the voltage and current drawn by the load. CAD designs of the telemetry
BoB were conducted using SolidWorks and the telemetry BoB is designed to mount directly into
the CEReS power rack. Unique modular tray designs were implemented to further the agnostic
and easy maintenance design, which allows for connections to be rewired and configured
depending on thruster selection. This required modular tray design places limitations on the overall
mounting design dimensions, which required the analysis of thermal and stress calculations due to
the tight operating dimensions of the circuitry. Figure X. depicts the SolidWorks model of the final
telemetry BoB design, in which the mounting trays are visible. The thermal and stress analyses
were performed using a simplified version of the model, where the simulations required the
removal of the telemetry BoB plexiglass side panels for improved airflow and the removal of the
additional front 8020 aluminum bars for additional usable tray space. These simplified and ideally
assumed simulations can be found in Appendix C; the stress-strain analysis proved that the singular
front plexiglass panel and removed 8020 bars were still capable of supporting the BoB’s weight.
The thermal analyses also verified the airflow which involved a fully modeled fan pictured in the
final layout section.

Figure 7. Telemetry BoB CAD Model

14
 Table 2. Power Supply Information

CEReS Power Supply Characteristics

Power Power Power Power Power Power Power
Specs
Supply 1 Supply 2 Supply 3 Supply 4 Supply 5 Supply 6 Supply 7
EMS 60-18- DCS 300-3.5 EMS 60-40- EMS 60-18- EMS 30-20- GEN 600- GEN 600-
Model
1-D-10T 2-D 1-D-10T 1-D 5.5-1P230 5.5-1P230
Watts [W] 1000 1050 2500 1000 600 3300 3300
Max Volts
60 300 60 60 30 600 600
[VDC]
Max Amps [A] 18 3.5 40 18 20 5.5 5.5
Single-
Phase Single Single Single Single Single Single
Triple
200-250, 100-
Volts [VAC] 105-125 190-253 105-125 105-125 170-265 170-265
130
Amps [A] 11.3 10-20 28.5-16.5 11.3 11.3 23 23
Frequency [Hz] 47-63 47-63 47-63 47-63 47-63 47-63 47-63
Remote Analog
Yes Yes Yes Yes Yes Yes Yes
Programming

Each modular tray is designed for a specific voltage and amperage rating such that power
supplies can be swapped out within the rating limitation. For instance, the EMS 60-40 power
supply is capable of operating and supplying 40A, which would be beyond the current sensing
ranges of certain methodologies. This is where the individual tray ratings are inciteful as future
interactions and modifications will be able to reference these requirements. This was a critical
design driver and improvement over the previous telemetry system; however, each existing power
supply will be connected to a specific rack on the modular BoB as shown below.

15
 Figure 8. Telemetry BoB Power Ratings

Each modular telemetry tray will include a voltage divider for load voltage draw
monitoring and a current sensor for load current draw monitoring. The current sensor is located on
the load side of the voltage divider to ensure that the measured current is the drawn current of the
load, since a small current will be drawn by the voltage divider. Both the current sensor and voltage
divider are configured to return an analog signal via BNC connection to the DAQ and LabVIEW
system for analysis. Furthermore, to allow for additional agnostic capabilities, and the current
sensor trade study, each tray was designed to house multiple current sensor types. For instance, the
EMS 60-40 supply has the shunt-based sensor installed, while the GEN 600-5.5 has both the
FluxGate and custom-built CQ330E sensors. Each of the trays have ample space for these
additional current sensors to be installed, while the middle of each tray is specially only for the
Verivolt FluxGate or shunt-based sensors. The Vektrex VCS40 was left free floating between the
trays for modular access and hollow cathode experiments. The overall dimensions of this BoB are
26in tall, 17in wide, and 5in deep with a usable tray area of 5in tall, 15in wide, and 5in deep. These
current sensors are discussed in the following sections. The below diagram details the schematical
layout of each tray.

16
 Figure 9. Circuitry Layout for Telemetry BoB Trays

2.1.2 Final Design

Referencing the determined dimensions, materials, and layout of the Telemetry BoB, the
final design and prototype offers the desired modularity with minor alterations. For each of the
seven trays the FluxGate and shunt-based current sensors are mounted directly in the middle of
the removable trays on DIN rails. Besides these are the other variants of the current sensors and
the voltage dividers. Both the custom-built CQ330E and voltage divider PCBs utilize plastic
standoffs to isolate the circuitry from other pieces of equipment. The below figures picture the
overall front views of both the entire Telemetry BoB and a single panel. The overall view shows
the modular and agnostic design where the inputs, outputs, power supply, and ground are all
attached identically and connect externally per DC power supply and EP device.

Figure 10. Telemetry BoB Tray

17
Figure 11. Telemetry BoB Prototype

18
2.2 Discharge Filter and Power Control BoB
2.2.1 CAD and Simulations
Discharge filters are specific to HETs due to significant challenges brought on by HET
discharge instability. These instabilities arise from plasma oscillations that develop during initial
startup and operation. For instance, discharge current oscillation can range from 10-100 kHz and
cause increasing damage and erosion along the channel and beam divergence. To counteract this
instability, a combination of a variable resistor, capacitor, and inductor and utilized as protective
circuitry. This combination not only mitigates these instabilities but protects expensive hardware,
and the HET itself. Figure X. depicts the prior layout of the discharge filter, which was utilized for
its retrofit within the new BoB. Figure X. Shows the CAD model for the Discharge Filter/ Power
Control BoB where the protective circuitry and power routing subsystems are located. Thermal
analyses were performed using SolidWorks which resulted in the conceptual addition of two fan
ports and additional height improvements, which allows the cooling of these components. The
additional height also allows for future modifications and components to be added if ALPE or
CEReS were to receive new equipment or EP devices. The overall dimensions of this Bob are 10in
tall, 17in wide, and 27in deep.

Figure 12. Prior Discharge System Arrangement

19
Figure 13. CAD Model of Discharge Filter & Power Control BoB

20
2.2.2 Final Design
The discharge filter prototype was completed by retrofitting the original circuit layout,
components, and inputs while expanding upon the safety and modularity of the BoB. Referencing
the CAD model and conceptual plans, this was done by creating isolated plexiglass sections where
the induvial discharge filter components were installed and then taped over with dielectric and
Kapton tape. An easily accessible plexiglass door was installed where an isolated plexiglass shield
is attached on standoffs for safety between the discharge components and inputs. The inputs are
retrofitted from the old design and are updated with higher amperage bananas to prevent burning
and damage. Each input is labeled and validated through installation tests. Neither the fans or
power control were implemented due to the phasing availability on ALPE’s breaker supplies;
though, the plexiglass sections were still installed for their future installation. The below figures
show the final prototype design.

Figure 14. Discharge Filter Banana Connections

21
 Figure 15. Discharge Filter Prototype
2.3 Current & Voltage Sensors
2.3.1 Telemetry Current Sensor Trade Study
Four options were considered for current sensing in each telemetry tray. Custom-built, Flux
Gate, Shunt-based and Hall current sensors were investigated for their agnostic capabilities and
tradeoffs. Input power to these devices comes from the EML30 AC-DC power supply with a
maximum 5V output. This trade study was conducted to not only compare the various current
sensor methodologies for their accuracies, costs, and operational performance, but also for the
agnostic and modular capabilities. For instance, the modular and agnostic designs require a
seamless and relatively easy configuration between thruster selection. This happens through the
user routing each input into the desired location [8-10].
[Link] Custom-Built CQ330E Sensors
The CQ330E Hall sensor (IC1) is integrated with a low pass filter (R1 & C2) and LED indicator
(LED1) when active for high-speed current sensing. The maximum input current capability is 20.5
A. This method of current sensing is attractive due to its low cost, but the 20 A maximum input
current is not large enough to accommodate requirements for all of the trays. A PCB design was
developed for testing the CQ330. Figure 12. displays the custom-built circuit schematic using
screw terminals for the power input from the EML30 AC-DC power supply, and wire to PCB
terminals for signal input and output. Figure 14. displays the custom-built PCBs with dimensions
of 50mm x 50mm and a thickness of 1.6mm, where the selected material is FR4-Stanard TG 135-
140. This design also encompasses M3 compliant mounting holes along the corners for standoff
capabilities to isolate the circuitry from potential shortages. The input signal from the back of each
power supply enters J1 and exits on J2; between these lies the CQ330E Hall sensor chip which is
powered by an external DC power supply entering on J3. The output signal passes through the
aforementioned low pass filter and outputs on J4 to a BNC bulkhead and directly to the DAQ.

22
Figure 16. Custom CQ330 Hall Sensor Circuit Schematic

Figure 17. Custom CQ330 Hall Sensor PCB

23
[Link] Flux Gate Sensors
The Verivolt IsoBlock I-FG-1c was selected because of its reasonable cost and high accuracy.
These flux gate sensors measure current through the inductance change in a saturable inductor
core. Control circuitry provides an input current to nullify the current generated through the
inductor core. This model has a 15A input range and a 10V output range, allowing for
measurements on most of the CEReS power supplies. These sensors are a single channel output
and input with wire terminals for inputting and outputting the desired measurements and current.
These sensors are mounted directly to DIN rails. This method provides high-quality, isolated
measurements with an accuracy of ±0.2% and a bandwidth of 100-200 kHz.

Figure 18. Technical Drawing of the Verivolt IsoBlock Sensor

Figure 19. FluxGate Methodology

24
[Link] Shunt-Based Sensors
The Verivolt IsoBlock I-ST-1c is a high-performance, shunt-based current sensor capable
of sensing 100A for the EMS 60-40 at an accuracy of 0.2%. This methodology has a 100A range
and 10V output range with a bandwidth of 100-200 kHz. In essence, the shunt methodology
measures the current flowing through the input conductor by placing a high-performance, low
impedance resistor along the primary current path. A galvanic isolation, which separates the input
and outputs supplies, isolates the primary and secondary sides. The input current is then measured
by amplifying the voltage across the shunt resistor before entering the DAQ. This method is much
more desirable for the EMS 60-40 due to its relatively high current transient. Similar to the
FluxGate model, this sensor is also mounted directly to DIN rails.

Figure 20. Shunt-Based Methodology

[Link] Hall Current Sensors

Hall current sensors work by outputting a voltage generated that is proportional to the Hall
effect current generated in a magnetic core. The VEKTREX VCS40 produces an output voltage
depending on the magnetic field, which allows for the current to be measured. The selected
VEKTREX device for testing has a maximum 40 A measurement range, allowing for within-
tolerance current sensing for six of the seven channels. There is a ±0.5% accuracy drift as
temperature and amperage increases which requires optimal ventilation in the BoBs. A thermal
model was created to analyze the potential thermal conditions in the BoB under load conditions.
The VCS has an inboard BNC connection to reduce total system noise and allow for easier
connection to the DAQ. The maximum bandwidth of these magnetically isolated current sensors
is 650 kHz, which also for high-precision measurements to be taken in conjunction with a Pearson
Coil.

25
Figure 21. Technical Drawing of the VEXTREX VCS40

Figure 22. Hall Effect Methodology

26
[Link] Trade Study Results
A preliminary trade study was performed on all current sensing options before tests were
performed on the components considered. According to the preliminary results, the Verivolt
IsoBlock I-FG-1c current sensor is the most optimal choice. This is due to its relatively cost-
effective design, with desired accuracy and range. It is important to note though that the Vektrex
VCS40 is capable of operating at higher bandwidths than the other sensor options. The custom-
built CQ330E is also a viable choice for small scale projects and academic institutions, yet its
range is not as suitable for higher power experiments.

Table 3. Trade Study of Current Sensor Methodology

Current Sensor Trade Study (Lower Numbers Indicate Better)

Accura Averaged
Ease of Use Range cy Cost Metric
Relative Importance 0.1 0.2 0.4 0.3 1
Options
CQ330 Hall Sensor 3 4 2 1 2.2
Shunt Resistor 4 2 4 2 3
Verivolt IsoBlock Sensor 2 1 1 3 1.7
VEKTREX VCS Hall Sensors 1 3 3 4 3.1

2.3.2 Telemetry Voltage Divider Design

Voltage dividers are used to adjust the level of signals by stepping down the incoming
voltage before measurement. An array of traditional voltage dividers was implemented to allow
for safe measurement of the power supply output voltages with National Instrument’s USB-6229
DAQ unit. Fundamentally, voltage dividers work through the technique of splitting a high voltage
into two separate voltages using two in-series resistors. The voltage divider designs chose to
implement 2010, thick film series resistors for easier soldering and a higher wattage rating for
improved protection. These resistors are governed by the following formula:

=
+
Using this formula, the voltage divider values and designs were determined for each of the
seven power supplies. Due the transient nature of EP, further protection was added to these designs
by the implementation of fast burning fuses from the 216 Series from Littlefuse and the Fast-
Acting Fuse 0ADBP and 0ADAP Series from Bel Fuse. These fuses were chosen due to their
opening times and voltage ratings. Each design consists of a compact ceramic body, cartridge
design which allows for seamless integration for supplementary protection from transient jumps

27
in amperage experienced during EP operations. These fuses were calculated based on Figures ??
which detail the opening time (blowing) of the fuses based on amperage, and the formula =
1.25 where a transient jump of amperage is accounted based on the electrical characteristics of
each power supply. The below figures and tables reflect the decision-making processes used to
determine the fuse values and the chosen resistor values.

Figure 23. Average Time Current Curves of 216 Series (left) & 0ADBP & 0ADAP Series (right)
Fuses

Table 4. Voltage Divider Design 1

Resistor Design 1
Power R1 R1 Model # R2 Rating R2 Model #
Supply Rating [k]
[k]
EMS 60-40 125 RMCF2010FT124K 10 RMCS2010FT10K0
EMS 30-20 60 RMCF2010FT60K4 10 RMCS2010FT10K0
EMS 60-18 125 RMCF2010FT124K 10 RMCS2010FT10K0
DCS 300-3.5 350 RK73H2HTTE3483F 5 RMCS2010FT4K99
GEN 600- 600 AC2010FK- 5 RMCS2010FT4K99
5.5 07619KL

28
 Table 5. Voltage Divider Design 2

Resistor Design 2
Power R1 R1 Model # R2 Rating R2 Model #
Supply Rating [k]
[k]
EMS 60-40 125 RMCF2010FT124K 10 RMCS2010FT10K0
EMS 30-20 60 RMCF2010FT60K4 10 RMCS2010FT10K0
EMS 60-18 125 RMCF2010FT124K 10 RMCS2010FT10K0
DCS 300- 350 RK73H2HTTE3483F 5 RMCS2010FT4K99
3.5
GEN 600- 600 AC2010FK- 5 RMCS2010FT4K99
5.5 07619KL

Table 6. Voltage Divider Fuse Values

Fuse Design
Power Supply Rating [A] Model #
EMS 60-40 16 [Link]
EMS 30-20 8 [Link]
EMS 60-18 8 [Link]
DCS 300-3.5 1.6 0ADAP1600-RE
GEN 600-5.5 2.5 0ADBP2500-RE

Utilizing the above resistor and fuse values, the voltage divider circuit and PCB schematics
were modeled using EasyEDA, a free PCB and circuit design platform. The input and output
signals utilize Buchanan series 2 position wire to board terminal blocks, allowing for properly
gauged wires to be seamlessly connected between the telemetry and DAQ systems. The designed
PCBs are 54mm x 36mm with a thickness of 1.6mm, where the selected material is FR4-Stanard
TG 135-140. This design also encompasses M3 compliant mounting holes along the corners for
standoff capabilities to isolate the circuitry from potential shortages. On the voltage divider
schematic, J1 represents the input coming out of the back of each power supply and is attached to
the PCB through wire terminals. VCC is the signal path as it runs through F1, the fuse, and the
voltage divider R1 and R2 before outputting in J2 on the same model wire terminal block. These
outputs are then connected and soldered to BNC bulkheads and connect directly to the CEReS
DAQ. The following figures detail the circuit schematic used to design the system, as well as the
final voltage divider PCB layout.

29
Figure 24. Voltage Divider Schematic

Figure 25. Voltage Divider PCB Layout

30
2.4 Overall Design and Layout
Referencing the below figure, installation of all components is to be arranged in accordance
with the discharge filter and power control being on the bottom and the Telemetry BoB being
directly above those. The discharge filter and power control BoB is on the bottom due to cable
management of the breaker inputs running along the floor. This in turn frees up additional space
for rearrangement of the prior CEReS power rack setup. The power supplies are arranged from
high amperage to lowest, where the EMS 60-40 is accessed and mounted through the top tray, and
the final GEN 600-5.5 is accessed and mounted through the bottom tray. Since the Telemetry BoB
is only 5in deep, the already installed power supplies and equipment do not need to be adjusted as
the space between them is solely for running wires. Additionally, the modeled and simulated fans
are attached both within and above the arrangement to provide optimal ventilation and cooling for
maximum thermal performance.

Figure 26. Rack Arrangement

31
3. Remote Control and Monitoring
The remote control and monitoring system includes both serial and analog connections to
CEReS equipment and is fully integrated with a LabVIEW virtual instrument. The equipment
being remotely monitored and controlled includes two ALICAT mass flow controllers, three gate
valves, and seven power supplies. The serial connections include the two mass flow controllers; a
conceptual design to serially control two Lambda GEN-model power supplies is also discussed.
The analog connections include all seven power supplies and three gate valves; a conceptual design
to analog control of three Lambda EMS-model power supplies is also discussed. The remote
control and monitoring system is integrated with a LabVIEW virtual instrument for maximum
functionality and ease of use. Future recommendations for the remote control and monitoring
system are discussed at the end of this section.
3.1 Serial Connections
The serial connections include the two ALICAT mass flow controllers, allowing for remote
control and monitoring of gas properties. The mass flow controllers are connected via a custom
RS-232 cable to a computer COM port, from which the LabVIEW program communicates serially
using ALICAT instrument drivers. A conceptual design to serially control two Lambda GEN-
model power supplies is also discussed, including integration with LabVIEW using Lambda EMI
instrument drivers.
3.1.1 ALICAT Mass Flow Controllers

Figure 27. ALICAT Mass Flow Controller

The ALICAT mass flow controllers are connected to the computer via a custom RS-232
cable to a computer COM port, from which the LabVIEW program communicates to the
controllers using ALICAT instrument drivers. This serial connection allows for the remote
monitoring of controller data and the control of various mass flow controller features. Users of the
LabVIEW program are able to monitor pressure, temperature, volumetric flow, and mass flow of
the gas mixtures being used in experiments through both LabVIEW indicators and plots. The
LabVIEW program also provides data logging capability for the mass flow controllers; all
monitorable parameters can be logged. The LabVIEW program allows for the control of gas
selection, gas setpoints, sample intervals, and the creation of programmable gas mixtures. See
section 3.3 for more details on the LabVIEW virtual instrument design.
32
3.1.2 Lambda GEN Power Supplies

Figure 28. Lambda GEN-model Power Supply Unit

Due to the project timeline's time constraint, the two Lambda GEN-model power supplies
were not set up for serial control via the lab computer. However, the LabVIEW program was
designed in a way that allows for easy adaptation of the Lambda GEN-model power supply drivers.
The Lambda GEN-model LabVIEW drivers were downloaded onto the CEReS lab computer and
included in the folder with the LabVIEW program. This makes future Lambda GEN-model power
supply adaptation convenient for continued development. See section 3.3 for more details on the
LabVIEW virtual instrument design to give an idea of how implementation can be accomplished.
3.2 Analog Connections

Figure 29. Analog Communication Flow Chart

33
 The analog connections made for remote monitoring include all seven power supplies and
three gate valves. The analog time-averaged signal from the seven power supplies is captured using
the circuitry from the telemetry BoB and fed directly to the DAQ via BNC connections. The input
signal to each of the three gate valve relays is fed to a voltage divider to step down the voltage,
and the output to the voltage divider is fed directly into the DAQ for measurement. A conceptual
design for the analog control of the three Lambda EMS-model power supplies is also discussed
with Arduino Mega 2560 Rev3 microcontroller implementation.
3.2.1 Analog Remote Monitoring
The analog monitoring system includes all seven power supplies and three gate valves. The
power supplies are monitored via analog signal from the telemetry BoB and the gate valves are
monitored in LabVIEW via analog signal from the gate valve relays. These signals are sent to the
DAQ for measurement and integration with a LabVIEW virtual instrument.
[Link] Power Supplies

Figure 30. Power Supply Measurement Flow Chart

Each of the seven CEReS power supplies are monitored via circuitry from the telemetry
BoB. The power supplies are monitored via analog time-averaged voltage signals that are output
from the voltage dividers and current sensors in each tray of the BoB. The voltage signals from
the BoB are fed directly to the DAQ via BNC connections for monitoring and logging data.
Since the voltage and amperage output ranges of each power supply are different from one
another, each voltage divider and current sensor were designed to input a range of signals specific
to each power supply. However, each voltage divider and current sensor outputs a a specific full-
scale range (FSR), dependent on the sensor. The NI USB-6229 DAQ has a 16-bit resolution for
FSRs of ±200 mV, ±1 V, ±5 V, and ±10 V. Every voltage divider and current sensor was designed
to output a FSR near either ±5 V or ±10 V to the DAQ. This ensures that the sensor output FSRs
corresponds to a FSR of the DAQ, maximizing the digital resolution of the captured data.

34
 The designed sensor ranges and corresponding digital resolution after digitization are listed
below in Table 6. The “Output Full Scale Range” column represents the FSR of the output of the
corresponding sensor. The “DAQ Full Scale Range Used” column represents the FSR used by the
DAQ when receiving the FS output of the corresponding sensor. The “Digital Resolution Digital
Resolution of FS Sensor Input” column represents the digital resolution of the DAQ measurement
in terms of the sensor input signal at FS.

Table 7. Circuit Component Output Full Scale Range and Digital Resolution
Output Full Scale DAQ Full Scale Digital Resolution of
Range [V] Range Used FS Sensor Input
EMS 60-40 Voltage Divider 4.4444 ±5 V 1.000 mV
EMS 30-20 Voltage Divider 4.2857 ±5 V 0.534 mV
EMS 60-18 Voltage Divider 4.4444 ±5 V 1.000 mV
DCS 300-3.5 Voltage Divider 4.2254 ±5 V 5.400 mV
GEN 600-5.5 Voltage
4.9587 ±5 V 9.200 mV
Divider
CQ330E 4.1000 ±5 V 0.382 mA
FluxGate Verivolt IsoBlock 10.000 ±10 V 0.289 mA
Shunt Verivolt IsoBlock 10.000 ±10 V 1.500 mA
Vektrex VCS40 8.0000 ±10 V 0.763 mA

The digital signal output from the DAQ is used as input to the LabVIEW program so that
the user can monitor and log voltage and current data from each power supply. The LabVIEW
program allows any voltage divider and current sensor signal to be monitored or logged. See
section 3.3 for more details on the LabVIEW virtual instrument design.
[Link] Gate Valves

Figure 31. Gate Valve Status Measurement Circuitry

35
 The three gate valves used in the operation of the pump-down system of the CEReS
vacuum chamber are the roughing pump valve, the CTI-250F, and the CTI-10. Each gate valve is
open or closed by input from the user of CEReS by pressing its corresponding mechanical button
mounted onto the rack. The button acts as a switch and, after pressed, changes the state of a circuit
that consists of a 24 V power supply and an electrical relay. When this circuit’s state is changed,
the relay's state changes and, in turn, changes the state of another circuit that consists of the gate
valve and its power supply. Then, depending on the state of the gate valve circuit, the gate valve
will open or close.
To measure the status of each gate valve, the input signal to each gate valve relay is fed to
a voltage divider to step down the voltage for DAQ measurement. Each voltage divider includes a
10 kOhm resistor in series with a 2.64 kOhm resistor. The output from each voltage divider is 5
VDC when the mechanical button switch is closed ( = 24 V) and 0 VDC when the mechanical
button switch is open ( = 0 V). The voltage output of each voltage divider is then sent
directly to the DAQ. The digital signal output from the DAQ is used as input to the LabVIEW
program to monitor whether each valve is open or closed. When a valve is open, an indicator lights
up on the LabVIEW front panel.

3.2.2 Analog Remote Control

Figure 32. Lambda EMS-model Power Supply Units

The three Lambda EMS-model power supplies can only be communicated with through an
analog connection. Due to the time constraint of this project, an analog control system for the
Lambda EMS-model power supplies was not implemented. Instead, a conceptual design for the
analog control of the Lambda EMS-model power supplies is discussed. The control system
proposed in this section implements analog circuitry, an Arduino Mega 2560 Rev3 microcontroller,
and a LabVIEW virtual instrument for configuring control settings.

36
[Link] Circuit Design

Figure 33. Analog Control System Design Flow Chart

The conceptual design for control circuitry involves a LabVIEW program that
communicates with the DAQ to output DC signal to the Arduino. After receiving signal from the
DAQ, the Arduino transmits signals to the power supplies to configure them as specified by the
user. The output signals from the Arduino Mega 2560 Rev3 are pulse-width modulated (PWM)
signals and must be converted to a DC signal before it is received by the power supply. Conversion
from PWM to DC signal can be performed with a lowpass filter with negligible error. Simulations
were carried out for PWM-fed first order lowpass filters using LTSpice and MATLAB. A bill of
materials is discussed for first order lowpass filters, as well as a brief discussion of higher-order
filters [11]. This design requires two wire connections per power supply, one for current control
and one for voltage control. Since there are three EMS-model power supplies, six wire connections
are needed between the Arduino and the power supplies. This also means six wire connections are
needed between the DAQ and the Arduino. The Arduino Mega 2560 Rev3 has 15 PWM outputs
and 16 analog inputs. Thus, there are enough electrical connections on the Arduino for this
application, confirming the use of this design [11].

37
 Figure 34. Pulse-width Modulated Digital Signal Duty Cycle
A PWM signal is a binary signal that is quickly cycled between on and off states to simulate
an analog signal. The ratio between the time the signal is in the on state to the time the signal is in
the off state is called the “duty cycle”. The time between the start of each “on” state is constant
and known as the “cycle period”; the inverse is called the “cycle frequency”. The Arduino Mega
2560 Rev3 features PWM output with cycle frequencies up to 31,250 Hz and duty cycles anywhere
from zero to one hundred percent. The PWM output voltage of the “on” state is 5 V and the “off”
state is 0 V [11]. Using the Arduino for control applications as simple as illuminating an LED, a
PWM voltage signal is “good enough” at simulating an analog signal. This is because the PWM
cycle frequency is fast enough that the human eye cannot detect a difference between PWM input
and a DC analog input. Even though the PWM output of the Arduino is binary, and the LED is
turning completely on and completely off, the human eye cannot tell [11].
However, for the application of controlling power supply output for CEReS, a PWM
voltage signal is not sufficient. This is because the power supply control input directly correlates
with the power supply output. Thus, if the power supply control input is at full scale (the “on” state
of the PWM duty cycle), the output of the power supply will also be at full scale. So, if the power
supplies were directly fed a PWM input, the output of the power supplies would constantly be
cycling between full scale and zero. This is clearly not appropriate for the application of CEReS.

Figure 35. Pulse-with Modulated (left) to Analog Signal (right) conversion

38
 Instead, a DC voltage signal is needed to maintain a constant power supply output. Cheap
and easy conversion from PWM to DC signal can be performed with an analog lowpass filter.
Intuitively, the lowpass filter “smooths out” the signal to be more like a DC signal because it
attenuates high frequency oscillations. Due to its simplicity, a first order lowpass filter consisting
of a resistor and capacitor is investigated.

Figure 36. First Order Lowpass Filter

First order filters are well-known for having a tradeoff between settling time and peak-to-
peak ripple voltage. Based on the values of the resistor and capacitor in the lowpass filter circuit,
the settling time and ripple voltage changes. The maximum possible peak-to-peak ripple voltage
of the filter output is the case of a PWM duty cycle of 50%, given by the equation below, where
is the PWM cycle period and is the FS voltage of the PWM output.

=
4
The settling time increases as the time constant increases, as shown by the equation
below, where is the resistance of the resistor and is the capacitance of the capacitor.
4
= ln = ln

Thus, the larger the time constant, the smaller the ripple voltage and the larger the settling time.
If the settling time and ripple voltage are known, both the resistance of the resistor and the
capacitance of the capacitor can be determined by rearranging these two equations. Determining
the tolerable limits for the settling time and peak-to-peak voltage ripple of the power supply output
is necessary for determining the allowable limits for the settling time and peak-to-peak voltage
ripple of the lowpass filter output. However, determining the tolerable limits for settling time and
peak-to-peak voltage ripple of the power supply output is beyond the scope of this project, since
doing so would require an extensive knowledge and background on CEReS experiments.

39
[Link] Simulations
For the EMS-model power supplies in the CEReS rack, the FS voltage for voltage control
input is 5 V and the FS voltage for current control input is 100 mV. Thus, for current control, the
voltage output of the Arduino needs to be passed through a voltage divider to step down the voltage
from 5 V FS to 100 mV FS. A second resistor is used for the current control circuit for this purpose.

Figure 37. Simulated LTSpice Circuits

Even though tolerable limits for settling time and ripple voltage of the power supply
outputs were not determined, simulations were carried out for PWM-fed first order lowpass filters
using LTSpice. For these simulations, typical resistance and capacitance values as researched
online were used to investigate the power supply output response of each circuit design. This
enables future groups that work on this project to get an idea for the types of values they might see
when dealing with typical resistances and capacitances of first-order filter designs, serving as a
starting point for further analysis.
60 V FS Voltage Supply 40 A FS Current Supply

Figure 38. Simulated LTSpice Results of Voltage Supply (left) and Current Supply (right)
40
 Figure 31 shows the simulation results. The ripple voltage tables list the ripple voltage or
ripple current of the power supply output, based on a 60V FS voltage supply and a 40A FS current
supply. All simulations were carried out assuming a PWM frequency of 31,250 Hz and a duty cycle
of 50%. The largest PWM frequency of 31,250 Hz was chosen for analysis because this is what
would likely be chosen for implementation due to a high frequency corresponding to a low ripple
voltage. A duty cycle of 50% was chosen because this offers a “worst-case-scenario" of the largest
ripple voltage for any chosen duty cycle.

[Link] Bill of Materials

In total, there are three lowpass filters needed for voltage control, three lowpass filters
needed for current control, and an Arduino Mega 2560 Rev3. An estimated bill of materials was
worked out for this system design as seen below. These prices are based on rough average market
rates for components at the time of this document. Thus, the design will cost around $70 to $90
depending on the exact components used.

Table 8. Analog Power Supply Control System Component Costs

Choices for R1 Choices for R2

1 kΩ 5 kΩ 10 kΩ 20.5 Ω 102.5 Ω 204.1 Ω
$6 per 100 ct $6 per 100 ct $6 per 100 ct $6 per 100 ct $6 per 100 ct $6 per 100 ct
Choices for Capacitors Arduino Mega
1 µF 10 µF 100 µF 1 mF 10 mF 2560 Rev3
$8 per 100ct $8 per 150ct $6 per 10ct $5 per 10ct $7 per 5ct $50

[Link] Additional Considerations

If an acceptable ripple and settling time cannot be established with a first order lowpass
filter, other circuit designs need to be investigated for PWM to DC conversion. The next most
simple design, after a first order lowpass filter, is the use of a higher order filter such as a second
order RLC filter or a third order Butterworth filter. Use of a higher order filter has several
advantages over a first order filter, such as reduced ripple and faster attenuation of high frequency
oscillations. The downsides of higher order filters are an increased complexity, cost, and a greater
sensitivity to component variation. Due to the large number of factors to be considered, the
investigation of higher order filters is left for future work on this project.
A more complex option for PWM to DC conversion is the use of a PWM rectifier. A PWM
rectifier is a more elaborate way of converting PWM signal to DC signal and can be used in series
with a lowpass filter for further improvements on ripple reduction. The downsides of implementing
a rectifier circuit are an increased design complexity and cost. Again, due to the large number of
factors to be considered, the investigation of rectifiers is left for future work on this project.

41
3.3 LabVIEW Virtual Instrument
A virtual instrument was created in LabVIEW to remotely monitor and control equipment
in CEReS, including the analog DAQ inputs, thermocouples, the ALICAT mass flow controllers,
and oscilloscopes. The LabVIEW program is designed in a user-friendly way so that researchers
that are new to CEReS can easily start working with the program. The user-interface is intuitive,
and the block diagram structure is organized and straight-forward to diagnose. The program
includes all necessary controls and indicators in a single program, allowing for complete
monitoring and control of the entire system on a single screen.
3.3.1 Front Panel Design
The program's front panel has four tabs of controls and indicators: Telemetry, Control,
Configure, and Instructions. The Telemetry panel is the main panel that is used for telemetry
purposes, the Control panel is the panel that is used for controlling the mass flow controllers and
power supplies, the Configure panel is used for configuring the settings needed for data acquisition
and telemetry, and the Instructions panel is used as a quick-start instruction guide to running the
program for new users.

Figure 39. LabVIEW Telemetry Panel

The Telemetry panel is the main panel used for telemetry purposes. The Telemetry panel
consists of 32 indicators and four charts. The design is extremely modular such that the user can
select each indicator and plot display from the corresponding ring controls, which uses every DAQ
channel specified in the Configure tab as a selection option. Each of the four charts can plot up to
four plots at the same time. This means 32 DAQ channels can be monitored from indicators and
16 DAQ channels from charts, making up to 48 DAQ channels able to be monitored at any given
time. The Telemetry begins operation after the program has been configured via the “Configure”
button on the Configure tab and then the “AUTORECORD” button is selected.

42
 Figure 40. LabVIEW Control Panel
The Control panel is used for controlling mass flow controllers and power supplies. It
consists of a section on the left for controlling each of the seven power supplies and a section on
the right for controlling each of the two mass flow controllers. At of the time this document was
written, the telemetry system is not set up for the remote-control of power supplies and so the
Control tab reflects this by indicating “N/A” under the “Set” section of each power supply and
"Power” buttons for each power supply are also nonfunctional. However, the Control tab does
display the load sensed at each power supply in the “Read” section of each power supply, where
the respective DAQ channels are configured on the Configure tab.
The mass flow controllers are remotely operated via the right-hand side of the Control tab.
The program allows the user to set the Setpoint and Gas of each mass flow controller, select the
interval for data capture, easily create gas mixtures, and select which relevant physical parameters
to log or plot. The “Take Data” button must be activated for the mass flow controller data to be
acquired. The “Log Data” button must be activated for the data in the “Data Logging” cluster to
be logged. If enabled, the two graphs in the bottom right corner automatically plot the parameters
checked in the “Graphing Options” control.

43
 Figure 41. LabVIEW Configure Panel
The Configure panel is used for configuring the settings needed for data acquisition and
telemetry. The lefthand side of the panel has a list of DAQ inputs that are configurable and the
righthand side of the panel has sample interval, file path, scope, power supply channel selection,
and an error code display. Once all controls are configured as the user desires, the user presses the
“Configure” button to store all settings and initialize the instrumentation drivers. If there is no
program error, the “Status” indicator illuminates.
On the lefthand side of the Configure panel, the DAQ input list is configurable with options
for instrument type, channel number, name of input, measured value, gain, offset, minimum
allowed value, maximum allowed value, alarm configuration, data recording, DAQ range
selection, and thermocouple type. The "Instrument” ring controls are selectable for Analog or
Thermocouple inputs. The “Channel” controls are the corresponding DAQ channel for a given
DAQ input. The “Name” controls are the user-defined names for each DAQ input. The
“Measurement” controls are the measurement readings from the DAQ. The “Gain” and “Offset”
controls are the linear gain and offsets applied to each DAQ input after the measurement is taken
and before it is displayed. The “Min” and “Max” controls are the minimum and maximum
allowable values measured before an alarm is tripped. The “Alarm” and “Record” controls enable
alarms and data recording for each DAQ input. The “AI Range” controls define the voltage range
for the DAQ input and is necessary for DAQ inputs specified as “Analog” on the “Instrument”
control. The “TC Type” controls are the type of thermocouple being used as the DAQ input and
are necessary for DAQ inputs specified as “Thermocouple” on the “Instrument” control.

44
On the righthand side of Configure panel, there is a Com Port input for the Agilent 34980A,
controls to change the sample intervals for both analog and thermocouple instruments, file paths
for the auto-record and single capture data files, oscilloscope configuration settings, power supply
channel selection, and an error code display. The oscilloscope configuration settings have options
for configuring up to three scopes with four channels per scope. The “Type” controls allow for
selectable data type, including mean, root mean square, peak to peak, amplitude, and standard
deviation. The instructions panel provides a detailed, step-by-step guide on using the program.
This allows new users to easily learn to use this virtual instrument.

Figure 42. LabVIEW Instructions Panel

3.3.2 Block Diagram Structure

The block diagram is structured in a format that best makes sense for the application of
telemetry and remote control. The program uses a sequence structure as its outer-most structure,
with the first frame in the sequence being a frame that allocates the memory necessary for the
controls and indicators located on the front panel. The second frame is the main frame of the
program in which most of the data processes take place. It contains an event structure in which
there are two main events: one for configuration of all equipment and another for recording and
logging of data. The first event, configuration, is triggered by a value change of the ”Configure”
button on the Configure tab of the front panel and includes storing all data necessary for properly
configuring all equipment, after the user specifies the correct information. The second event,
recording and logging data, is triggered by a value change of the ”AUTORECORD” button on the
front panel and includes recording and logging all data specified during the previous configuration
event.

45
4. Sensor Benchmarking
4.1 Simulated Load Tests

Simulated load tests were performed on each component considered in the current sensor
trade study, as well as on the voltage dividers. For these tests, each component was separated
from the BoB and connected to an output power supply such that the accuracy and linearity of
each component could be evaluated. This was done by directly linking a simulated load to the
EMS 60-40 power supply through the setup detailed in the below figures. The load simulates
what the actual power supply and Telemetry design will perform during operation. The power
supply was set to a specified voltage, as the load equipment had a maximum wattage of 150W.
Throughout the tests the amperage on the load equipment was adjusted from 0A to 20A and the
computed data was analyzed using an external multimeter.

Figure 43. Experimental Load Setup

46
 Figure 44. Sensor Benchmarking: Current Sensor Load Tests

In this test, the Vervolt shunt based, Verivolt FluxGate, and Vektrex VCS40 demonstrated
linearity within 1%. The CQ330E struggled to demonstrate linearity. An overvoltage on the test
board is assumed to be the cause of the anomalous result. However, even before the threshold of
the anomalous voltage, the CQ330E was still performing at a level inferior to the other options.
Through this testing, it is recommended that the CQ330E not be used for current sensing, even
when resolution is theoretically higher than that of a FluxGate or VCS40. Even accounting for the
increased resolution, the overall accuracy is better. Three voltage dividers, all calibrated for use
with different input power supplies, were tested for their output when subject to test signals. The
voltage dividers showed a poor level of performance, with reliability, linearity, and accuracy
severely lacking in comparison to initial projections.

Figure 45. Sensor Benchmarking: Voltage Divider Load Tests

47
 Both voltage dividers for the EMS 60-18 and the EMS 60-40 experienced failures
presumed to involve the fuse, resulting in a decreasing output as input increased. The voltage
divider designed for use with the EMS 30-20 did have a roughly linear relationship, indicating
avoidance of a board failure, but the linearity was not good enough to warrant use in the telemetry
system. Moving forward, the first priority to reach full system functionality is to focus on the
voltage dividers while re-evaluating and reintegrating them back into the system.

4.2 H6 Hollow Cathode Tests

The Vektrex VCS40 was tested with operation of the H6 hollow cathode. The H6 hollow
cathode was designed at the University of Michigan for use with a high-power HET. During
cathode testing, both the VCS40 and Pearson Coil were used to collect current data. It is important
to note that the VCS40 collects coupled AC/DC measurements, while the Pearson Coil only
provides AC measurements. In this graph, the total current reported by the Pearson coil includes
the DC component recorded by the VCS40. The phase difference observed is a result of how the
Pearson coil is placed closer to the load than the VCS40. The frequency response is almost identical
within the shared bandwidth of both sensors. This data is then normalized to observe the time
response. The transient, high-speed response between the Vektrex VCS40 and the benchmarked
Pearson Coil can be seen below.

Figure 46. H6 Testing

48
Figure 47. H6 Testing: VCS40 vs Pearson Coil, Non-Normalized, Magnetized

Figure 48. H6 Testing: VCS40 vs Pearson Coil, Normalized, Magnetized

Figure 49. H6 Testing: VCS40 vs Pearson Coil, Normalized, Nonmagnetized

49
 In the normalized figure, it is seen how the Pearson coil responds more quickly to changes
in current than the VCS40. This testing suggests that for high-precision, high-frequency signals,
the Pearson coil should be used in conjunction with the VCS40 or Verivolt coupled AC/DC sensors.
When the emitting cathode is not magnetized, higher frequency oscillations are more prevalent in
the current trace. It can easily be seen here how the VCS40 struggles to capture oscillations above
650kHz, its cutoff frequency. Future testing should include a comparison of the VCS40 and
Verivolt current sensors to investigate how the smaller bandwidth of the Verivolt will impact
performance in a direct comparison to the larger bandwidth VCS40.

5. Societal Impacts

This telemetry system is designed to enable fast turnaround testing for EP devices, enabling
faster development of propulsion systems that will have a great impact on humanity’s ability to
venture further into the cosmos. The testing practices created here can be used as a guide for future
engineers in the field to test with efficiency and safety. Power supply and telemetry systems used
in EP testing can output dangerous levels of voltage and current. In the design of this system, great
care was taken to ensure that fast turnaround testing could be performed while maintaining a
rigorous level of safety. This includes ensuring all components are rated for the maximum expected
transient spikes and having all electrical contacts encased in plexiglass. Fragile systems, such as
the DAQ, are protected by quick-burn fuses for protection. Many thermal simulations were also
performed during the design phase to allow ample airflow to cool the circuitry and decrease the
risk of possible overheating. These protective systems allow researchers and students to operate
these devices without being actively exposed to the high voltages and currents, or risks of fires and
damage. These elements create a safe design that serves as a model for those developing other EP
telemetry systems in the future.

50
6. Conclusion

Design of the telemetry and remote-control systems produced hardware and software
capable of integrating and operating with CEReS’ subsystems during agnostic thruster
experimentation was completed, with limited testing on the platform’s capabilities performed. The
revamped telemetry BoB descoped from the original goal which included separating the time-
averaged and time-resolved circuitry for measurement of low-frequency and high-frequency
signals. Due to cost and usability considerations, modular trays for each power supply were
constructed that house an AC/DC coupled current sensor and voltage divider, each connected to a
main DAQ unit. The rating-based design of each tray allows for the interchange of power supplies
and various EP devices. When HETs are tested, a selectable discharge filter can be optionally
connected on the load side of any of the trays in a switchboard fashion. The main goal of
developing a modular, agnostic telemetry system that enables quick turnaround testing on different
EP devices was met with this design.
Initial validation testing included the individual load testing of each considered current
sensor and voltage divider, with the current sensors providing an adequate measurement.
Integrated testing with the high-power, H6 hollow cathode showed that when high frequency
measurements are made. It was determined that the pairing of a high-speed Pearson coil with a
coupled AC/DC unit provides the best measurement accuracy. In the case of the voltage dividers,
their performance was determined to be wholly unsatisfactory. Quick action must be taken to
ensure system functionality due to the failure of the voltage dividers.

7. Future Recommendations

Furthermore, validating the installation of the telemetry BoB will require the operation of
a low-power HET (WHTms-44) to confirm agnostic capabilities. A comparison of the VCS40 and
Verivolt current sensors to investigate how the smaller bandwidth of the Verivolt will impact
performance in a direct comparison to the larger bandwidth VCS40 will be necessary to improve
the usage guidelines for the BoB system. Turnaround time on configuration swap will be tested
and the accuracy of different current measuring options. This testing will enable the development
of a guide for using the system, as well as producing performance metrics for the telemetry
hardware. Future work will require the validation of each telemetry circuit tray through simulated
operation of the EMS 60-18 and DCS 300-3.5 power supplies. These power supplies will need to
be utilized due to their unique input and output parameters. Future work will also include the
continuing development of the LabVIEW control suite to include the implementation of a power
supply remote control. Benchmarking details for this system are outlined in section 3.2.2.
Additionally, the BRAX pressure sensor is to be included in future work, as a [Link] program was
developed by PhD student Margaret Mooney. The LabVIEW program was designed to easily
implement both power supply remote control and BRAX monitoring adaptations.

51
 Acknowledgements

This project's completion could not have been possible without the assistance and guidance
of PhD student Margaret Mooney and faculty advisor Dr. Kristina Lemmer. Margaret Mooney
provided exceptional insight during routine progress meetings and guided this project to
completion and submission to the 38th International Electric Propulsion Conference. We thank
Margaret Mooney for her confidence in us and Dr. Kristina Lemmer for letting us use her
Aerospace Laboratory for Plasma Experiments.
We would also like to thank our sponsors and funding sources from Western Michigan
University’s Lee Honors College, NASA’s Joint Advanced Propulsion Institute, and NASA’s Space
Technologies Graduate Research Opportunities Fellowship (80NSSC20K1179). The financial
backing of these organizations and opportunities made not only the trade study and overall project
obtainable but allowed for this work to be published and presented at the 38 th edition of the
International Electric Propulsion Conference in Toulouse, France.

52
 References

[1] D. M. Goebel, I. Katz, Fundamentals of Electric Propulsion: Ion and Hall Thrusters,
Pasadena, California, USA: JPL Space and Technology Series, 2008.
[2] M. Baird, T. Kerber, R. McGee-Sinclair, K. Lemmer, “Plume Divergence and Discharge
Oscillations of an Accessible Low-Power Hall Effect Thruster,” Appl. Sci., vol. 11, no.
1973, pp. 1-4, Feb. 2021. doi: 10.3390/app11041973.
[3] M. Mooney, M. Baird, K. Lemmer, “Featherweight Heaterless Hollow Cathode
Characterization,” in 36th IEPC, Vienna, Austria, 2019, pp. 2-5.
[4] Mooney, M. and Lemmer, K. "The effect of probe separation distance on the analysis of
plasma waves in a cathode plasma," AIAA 2024-0923, AIAA SCITECH 2024 Forum,
Orlando, FL, January 2024.
[5] J. R. Wertz and W. J. Larson, eds., Space Mission Analysis and Design, Torrance,
California: Microcosm Press, 1999.
[6] J. R. Wertz and W. J. Larson, eds., Space Mission Analysis and Design, Torrance,
California: Microcosm Press, 1999.
[7] “The Psyche Mission is a Journey to a Unique Metal-Rich Asteroid Orbiting the Sun
Between Mars and Jupiter.” [Link]. [Link]
(accessed April 15, 2024).
[8] S. Gupta. “Current Sensing Technique Using Different Current
Sensors.” [Link]. [Link]
a-circuit-with-different-current-sensing-techniques (accessed Mar. 12, 2024).
[9] F. Carden, R. Jedlicka, R. Henry, Telemetry Systems Engineering, Boston, MA, USA:
Artech House, 2002.
[10] Z. Xingsong, D. Xiaogang, L. Xiaofeng, Z. Ruiming, “Design of Embedded Software
Architecture for Spacecraft Electric Propulsion System,” in Journal of Physics: Conf.
Series, in Volume Title 2497, 2023, pp. 1-7, doi: 10.1088/1742-6596/2497/1/012006.
[11] A. Levido. “PWM DAC Ripple Reduction.” [Link]. [Link]
/resources/quickbits/pwm-dac-ripple-reduction-2/ (accessed April 15, 2024).

53
Appendix A: Decision Matrix

54
 Appendix B: MATLAB Code for Table 6 Calculations

%% NI USB-6229 DAQ
FSR1 = 5;
FSR2 = 10;
Resolution = 2^16;
%% EMS 60-40
Vin = 60;
R1 = 125e3;
R2 = 10e3;
Vout = (R2/(R1 + R2))*Vin;
DigRes = ((FSR1/Resolution)/Vout)*Vin;
%% EMS 30-20
Vin = 30;
R1 = 60e3;
R2 = 10e3;
Vout = (R2/(R1 + R2))*Vin;
DigRes = ((FSR1/Resolution)/Vout)*Vin;
%% EMS 60-18
Vin = 60;
R1 = 125e3;
R2 = 10e3;
Vout = (R2/(R1 + R2))*Vin;
DigRes = ((FSR1/Resolution)/Vout)*Vin;
%% DCS 300-3.5
Vin = 300;
R1 = 350e3;
R2 = 5e3;
Vout = (R2/(R1 + R2))*Vin;
DigRes = ((FSR1/Resolution)/Vout)*Vin;
%% GEN 600-5.5
Vin = 600;
R1 = 600e3;
R2 = 5e3;
Vout = (R2/(R1 + R2))*Vin;
DigRes = ((FSR1/Resolution)/Vout)*Vin;
%% CQ330E (DCS and GEN) (20.5 A range 4.1 V output)
Ain = 20.5;
Vout = 4.1;
DigRes = ((FSR1/Resolution)/Vout)*Ain;
%% FluxGate Verivolt IsoBlock I-FG-1c (15 A range at 10 V output)
Ain = 15;
Vout = 10;
DigRes = ((FSR2/Resolution)/Vout)*Ain;
%% Shunt-Based Verivolt IsoBlock I-ST-1c (100 A range and 10 V output)
Ain = 100;
Vout = 10;
DigRes = ((FSR2/Resolution)/Vout)*Ain;
%% Vektrex VCS40 (40 A range and 8 V output)
Ain = 40;
Vout = 8;
DigRes = ((FSR2/Resolution)/Vout)*Ain;

55
Appendix C: SolidWorks Simulations

56
Appendix D: IEPC Abstract

57
58